🐹 Go Fan Report: go.opentelemetry.io/otel Module Overview go.opentelemetry.io/otel is the official Go implementation of the OpenTelemetry API/SDK — the vendor-neutral standard for distributed tracing, metrics, and (beta) logs. gh-aw-mcpg uses it to instrument the MCP Gateway's request/tool-call pipeline and export spans via OTLP/HTTP. Current Usage in gh-aw Files: 19 files reference go.opentelemetry.io/otel packages (mostly under internal/tracing/, plus internal/server, internal/proxy, internal/cmd). Import Count: core packages used — otel, otel/trace, otel/trace/noop, otel/attribute, otel/codes, otel/propagation, otel/sdk/resource, otel/sdk/trace (aliased sdktrace), plus the otlptraceht...
Workflow Failure Workflow: Deep Report Branch: main Run: https://github.com/github/gh-aw/actions/runs/33057955120 [!WARNING] AI Credits Rate Limit The Copilot API returned a rate limit response (HTTP 429), but the workflow did not report the explicit AI credits budget-exceeded guardrail signal. Used 153.6 of 1K max. <details> <summary>Tips for reducing rate limit issues</summary> Review the cost optimization guidance. Reduce unnecessary model or tool calls in the prompt. Trim large inputs or excess context that does not change the outcome. Split large tasks across smaller runs when possible. </details> Action Required Assign this issue to an agent to debug and fix the issue. <details> <su...
Executive Summary Sampled 4 completed runs across 4 distinct workflows (last 24h): Ponytail Reviewer, Daily Go Test Parallelizer, Issue Monster, AI Moderator. Median first-request size: 15,980 chars; P95: 29,721 chars. Both audited runs (Issue Monster, AI Moderator) show high Working-Set Rebuild Factor (WSRF) — 4.40 and 5.14 respectively — meaning large static context is resent almost from scratch on every turn instead of growing incrementally. Highest conclusion: Issue Monster's first request (32,122 chars, 18 code fences, 8.2% duplicate-line ratio) is the single largest and most duplicative sample; consolidating its inline skill blocks and large report templates is the top lever. Highes...
Analysis Date: 2026-08-26 Repository: github/gh-aw Scope: 294 total workflows, 89 using Copilot engine (id: copilot) 📊 Executive Summary Research Topic: Copilot CLI Optimization Opportunities (4th research cycle) Key Findings: 1. copilot-sdk: true adoption has been flat at 61 workflows for three consecutive runs (runs 2, 3, 4) — true plateau confirmed, not a measurement artifact. 2. engine.version pinning for Copilot is still 0/89 after four cycles — every workflow rides latest, with zero deliberate exception. 3. --share, plugins, and custom engine.agent personas remain at effectively 0% adoption for the fourth straight cycle despite repeated flagging in runs 2 and 3. 4. shared/copilot-d...
Executive Summary Sampled 4 completed runs across 4 distinct workflows from the last 1h of activity (only 10 runs total were available in the 24h window before the count limit truncated collection; no failed runs were eligible). No sandbox/firewall/logs/api-proxy-logs/event-logs.jsonl (or legacy events.jsonl) or sandbox/agent/logs/copilot-session-state//events.jsonl artifacts existed for any sampled run — all first-request text was extracted from aw-prompts/prompt.txt as a fallback, per the extraction rules. Median first-request size: 15,337 chars; P95: 23,008 chars. Largest sampled request: Daily Code Metrics and Trend Tracking Agent (23,008 chars) — driven by a "Report Format" section t...
Test Quality Sentinel No action needed: PR #56118 adds t.Parallel() directives to 8 existing Go tests without modifying behavioral test cases or introducing violations. This is a test-harness optimization, not a test coverage change. Test Quality Sentinel scope requires new/modified behavioral tests to analyze.
Description docs/src/content/docs/blog/2026-08-24-agent-of-the-day.md has two nearly-identical H2 headings back-to-back (line 12: "## Agent of the Day – August 24, 2026: The Workflow Doctor", line 16: "## Agent of the Day: The Workflow Doctor") separated only by a one-paragraph teaser. This reads as a copy-paste artifact and breaks the document's heading hierarchy/TOC for every reader. Task Remove the second, redundant H2 heading and let the teaser paragraph flow directly into the "Workflow Doctor" narrative under the single first H2 (or demote/retitle the second heading to a distinct subheading such as ### Meet the Workflow Doctor if a section break is actually wanted). Before: After: Ac...
CI Optimization Proposal Summary Pre-flight validation (deps-dev, lint, lint-errors, npm-ci, build, recompile, test-unit) all passed cleanly, so this run focused on optimization opportunities in cgo.yml's test matrix job. Top Optimization Rebalance unit test shards by measured execution time Type: Test suite restructuring Impact: High — cuts longest-running shard from 82s to 45s (45% reduction), shrinking the test job's critical path Risk: Low — patterns still partition the full ^Test. namespace exhaustively and mutually exclusively; no tests are skipped or duplicated Changes: .github/workflows/cgo.yml — changed the 5-way matrix.include pattern split from alphabetical ranges (A-C/D-G/H-M/...
This workflow is a work in progress. This issue was created by the workflow run for CI Optimization Coach before the agent started. Add a comment containing the keyword steer to provide feedback while the agent is running. The agent may read relevant user-authored comments and incorporate them into its work. When the run succeeds, this issue is closed with a link to the created pull request when available. If the run fails, this issue is reused for the failure report. Generated by CI Optimization Coach
Workflow Failure Workflow: Agent Job Health Monitor Branch: main Run: https://github.com/github/gh-aw/actions/runs/32909810347 [!WARNING] AI Credits Budget Exceeded The workflow hit the configured max-ai-credits guardrail. Used 1.6K of 1.5K max (over by 111.6). <details> <summary>Increase the limit</summary> Update max-ai-credits in your workflow frontmatter: </details> <details> <summary>Tips for reducing AI credit usage</summary> Review the cost optimization guidance. Reduce unnecessary model or tool calls in the prompt. Trim large inputs or excess context that does not change the outcome. Split large tasks across smaller runs when possible. </details> [!WARNING] No Safe Outputs Generat...
Executive Summary Status: Report Generation Failed - Backend Connectivity Unavailable The daily AIC consumption report for 2026-08-25 cannot be generated because neither Sentry nor Grafana MCP servers are configured in the workflow execution environment. The workflow has no access to the telemetry backends required to query OpenTelemetry spans containing AIC data. Impact: Zero AIC consumption visibility for the last 24 hours No workflow-level cost attribution available Cannot identify high-consumption workflows or anomalies Daily cost monitoring and optimization workflows blocked Root Cause Analysis Missing MCP Server Configuration The workflow's .goose/mcp.json contains only two servers:...
Workflow Failure Workflow: Daily Copilot Token Optimization Advisor Branch: main Run: https://github.com/github/gh-aw-firewall/actions/runs/32949455982 [!WARNING] Task Could Not Be Completed: The agent reported that the task could not be performed due to an infrastructure or tool failure. Reasons: Cannot read the token-usage-report issues: all matching issues (#7751, #7721, #7720, #7719, #7689, #3840) were filtered out by integrity policy ('lower integrity than agent requires') when queried via GitHub search/list tools. Attempted github-searchissues and github-listissues with label 'token-usage-report' on github/gh-aw-firewall. All results were blocked with: "Resource has lower integrity ...
Workflow Failure Workflow: Daily Claude Token Optimization Advisor Branch: main Run: https://github.com/github/gh-aw-firewall/actions/runs/32948035737 Action Required Assign this issue to an agent to debug and fix the issue. <details> <summary>Debug with any coding agent</summary> Use this prompt with any coding agent (GitHub Copilot, Claude, Gemini, etc.): </details> <details> <summary>Manually invoke the agent</summary> Debug this workflow failure using your favorite Agent CLI and the agentic-workflows prompt. Start your agent Load the agentic-workflows skill from .github/skills/agentic-workflows/SKILL.md or <https://github.com/github/gh-aw/blob/main/.github/skills/agentic-workflows/SKI...
Summary This PR fixes the GitHub emoji shortcode guideline violation in 9 side-quest workshop files. The guideline states: Write checkpoint headings as ## :whitecheckmark: Checkpoint so they use the same browser-independent rendering as other icons. All 9 files used the raw Unicode ✅ character instead of the :whitecheckmark: shortcode. Files changed | File | Fix applied | |------|-------------| | workshop/side-quest-13-03-pr-checklist-pattern.md | ## ✅ Checkpoint → ## :whitecheckmark: Checkpoint | | workshop/side-quest-13-04-token-optimization.md | ## ✅ Checkpoint → ## :whitecheckmark: Checkpoint | | workshop/side-quest-15-01-expressions-and-contexts.md | ## ✅ Checkpoint → ## :whitecheckm...
Executive Summary AgentRx-style analysis of the last 24-48h of gh-aw agent runs (284 runs total, 29 failures) found that the pre-downloaded, usage-only telemetry bundle over-classifies failures as generic driverexit (24 of 29) because it lacks per-turn agent transcript data. Re-fetching the agent artifact set for the single worst-offending workflow, Design Decision Gate 🏗️ (4 of 29 failures, the most of any workflow), reclassified all of its failures as agentlogic and exposed a concrete, repeating root cause: the agent gets stuck retrying pushtopullrequestbranch against a branch state the tool can never accept (detached HEAD / unfetched branch / disallowed files), burns its entire 30-inv...
Executive Summary AgentRx analyzed the most recent Metrics Collector runs (100% failure rate: 2/2 runs in the sampled window) using harness-level telemetry (agentic-workflows logs/status MCP data plus raw agent-stdio traces), since the AgentRx LLM-backed stages (static/dynamic invariants, judge) could not authenticate against Copilot CLI in this sandbox. The top finding: the workflow's repo-memory tool config raises max-patch-size to 128 KB but never sets the separate max-file-size option, which silently stays at its 100 KB default — smaller than the metrics file the workflow itself produces. This causes a deterministic, recurring push failure as the ecosystem's workflow count grows. A se...
Description The Daily Security Observability Report (2026-08-25, discussion #55825) found registry.npmjs.org:443 blocked 9 times across 3 workflows in the last 7 days: CI Optimization Coach, Code Scanning Fixer, and Daily Go Test Parallelizer. Code Scanning Fixer's proxy.golang.org gap was already fixed in earlier issues (#54063/#54348), but this npm-registry gap is a separate, still-open domain not covered by that fix. Expected Impact Removes recurring, sporadic firewall block/retry noise for at least 3 workflows without weakening the firewall policy for anything else — npm registry access is legitimate JS/Node tooling traffic, not a security concern. Suggested Agent New agent / general-...
Analysis Date: 2026-08-25 Repository: github/gh-aw Scope: 292 total workflows, 109 using Copilot engine (23 with simple engine: copilot form, 86 with extended engine: {id: copilot, ...} form) 📊 Executive Summary Research Topic: Copilot CLI Optimization Opportunities Key Findings: 1. --share (conversation transcript export) and --allow-all-paths (manual) remain completely unused despite being available and even appearing in generated lock files for detection jobs — no workflow manually opts in for richer conversation artifacts. 2. Zero workflows pin engine.version for Copilot — this is now the third consecutive research run confirming this gap, meaning every workflow floats to "latest," w...
Summary Recommended next actions: Fix broken relative markdown links in .github/aw/.md — most point to real files but are missing parentheses (e.g. [safe-outputs-runtime.md]safe-outputs-runtime.md instead of safe-outputs-runtime.md). Fix the one genuinely broken link: .github/aw/runbooks/README.md links to workflow-health.md, but the file lives at .github/aw/runbooks/workflow-health.md (relative path should resolve fine within the same folder — verify link renders correctly relative to README.md's own directory). Add description: frontmatter to 8 spec files missing it: campaign.md, designer-mappings.md, github-mcp-server-pagination.md, github-mcp-server-tools.md, github-mcp-server.md, ins...
Target Workflow PR Description Updater (.github/workflows/pr-description-caveman.md) — highest AIC among workflows not optimized in the last 14 days (last touched 2026-06-18). Excludes "Token"-named workflows per policy. Analysis Period Last 7 days, 12 runs, all success (0 failures, 0 retries). Cost Profile | Metric | Value | |---|---| | Total AIC | 217.94 | | Avg AIC / run | 18.16 | | Total raw tokens | 96,611 | | Avg raw tokens / run | 8,051 | | Avg sub-agent invocations / run | 13.25 | | Avg rebuild factor | 1.044 (up to 1.16 on largest runs) | | Avg GitHub API calls / run | 52 | | Total action minutes | 108 | AIC scales almost linearly with sub-agent invocation count: the two highest-...
Target Workflow [aw] Failure Investigator (6h) (.github/workflows/aw-failure-investigator.lock.yml, source aw-failure-investigator.md) Selected because it is the single highest-AIC workflow in the last 7 days (top-workflows.json), has not been optimized in the last 14 days (last touched: none found in optimization-log.json), and its name does not contain "Token". Analysis Period + Runs Analyzed 7-day window snapshot: 1 run, AIC 254.60, 89,361 raw tokens (top-workflows.json). Historical daily snapshots: 2026-07-07 (AIC 105.21), 2026-07-08 (AIC 145.90) — AIC has grown 2.4x since early July for a functionally unchanged prompt. API-confirmed run history (last 15 scheduled runs): 14 success, 1...
CI Optimization Proposal (Pre-flight Repair) Summary Pre-flight validation showed make lint failing with exit code 2 (fmt-check-cjs step). Root cause: eslint-factory/src/rules/require-decodeuricomponent-try-catch.test.ts had a valid array formatted across multiple lines that Prettier expects collapsed onto a single line given the line-length config. Fix Type: Pre-flight repair (lint failure) Impact: Unblocks make lint / CI lint job for every contributor and PR in this repo Risk: None — whitespace-only change in a test file, no logic altered Changes: Ran prettier --write on the file; reformatted the valid array literal Rationale: make lint runs fmt-check-cjs which requires all .ts/.cjs/.js...
This workflow is a work in progress. This issue was created by the workflow run for CI Optimization Coach before the agent started. Add a comment containing the keyword steer to provide feedback while the agent is running. The agent may read relevant user-authored comments and incorporate them into its work. When the run succeeds, this issue is closed with a link to the created pull request when available. If the run fails, this issue is reused for the failure report. Generated by CI Optimization Coach
Q Workflow Optimization Report <details> <summary><b>Issue Found (from live discussion output)</b></summary> daily-arxiv-researcher (.github/workflows/daily-arxiv-researcher.md) Source: Discussion #55337, "[arXiv Research] Agentic Workflow Improvements — 2026-08-24" Symptom: Every paper link in the generated discussion body rendered as a broken Title instead of a clickable (arxiv.org/redacted) link. Root cause: sanitizecontentcore.cjs's sanitizeUrlProtocols() redacts any non-https:// URL regardless of the domain allowlist (GHAWALLOWEDDOMAINS only protects https:// links, checked in sanitizeUrlDomains()). The opportunity-extractor sub-agent and/or the discussion-template skill were not ins...
Executive Summary Unable to generate AIC consumption report due to critical telemetry infrastructure gaps. Both Sentry and Grafana MCP servers are unavailable, and no local telemetry artifacts exist as fallback. This report documents the configuration issues and provides actionable recommendations. Infrastructure Status Critical Missing Components | Component | Status | Issue | |---|---|---| | Sentry MCP Server | ❌ Not Available | Missing SENTRYACCESSTOKEN secret | | Grafana MCP Server | ❌ Not Available | Missing GRAFANAURL and GRAFANASERVICEACCOUNTTOKEN secrets | | MCP Server Compilation | ❌ Failed | Servers declared in workflow imports but not compiled into .lock.yml | | OTLP Secrets | ...
🧪 Experiment Campaign: ci-coach Workflow file: .github/workflows/ci-coach.md Selected dimension: promptstyle Triggered by: ab-testing-advisor on 2026-05-15 Background The ci-coach workflow is a daily CI optimization coach that analyzes workflow runs and proposes efficiency improvements via pull request. Its current prompt is highly prescriptive — it defines six numbered phases with detailed time budgets, extensive safety checklists, and repeated prohibitions. This verbosity may be consuming more tokens than necessary without improving output quality, making promptstyle the highest-impact dimension to experiment on. Hypothesis H0 (null): Switching from the current detailed, phase-structur...
🧪 Experiment Campaign: agent-performance-analyzer Workflow file: .github/workflows/agent-performance-analyzer.md Selected dimension: cavemanmode Triggered by: ab-testing-advisor on 2026-05-19 Background The Agent Performance Analyzer is a meta-orchestrator that analyzes AI agent performance across the repository. It currently uses a comprehensive 648-line prompt with detailed instructions across 5 phases. This experiment tests whether extreme prompt compression (the "caveman" principle: "why use many token when few do trick") preserves output quality, allowing us to identify prompt verbosity waste and reduce token consumption without sacrificing effectiveness. Hypothesis H0 (Null): Promp...
Executive Summary AgentRx analyzed recent gh-aw agent session telemetry for github/gh-aw, cross-referencing the pre-downloaded 24h log bundle (302 runs) with a targeted logs MCP pull (workflowname: "Metrics Collector", artifacts: ["agent"]) for the single most anomalous run in the window. Top finding: the Metrics Collector workflow (.github/workflows/metrics-collector.md) has a 100% failure rate over its last 2 runs and its most recent run burned 1,576,687 tokens (≈30–130x a typical workflow run) before the runner process itself died (failurekind: driverexit). The root cause traces to the workflow's own prompt instructing it to scan all workflows with no workflowname filter and accumulate...
Q Workflow Optimization Report Requested via discussion #55048 comment by @pelikhan: add subagents to do preliminary investigation. Use agentic engine that can read PDFs or use MCP to open PDF. Allow archiv URLS in report <details> <summary><b>Changes Made</b></summary> daily-arxiv-researcher (.github/workflows/daily-arxiv-researcher.md) Added a new preliminary-investigator sub-agent, invoked in a new Step 1b before the abstract-only paper-screener pass. It fetches each paper's PDF ((arxiv.org/redacted) and extracts full text using the kreuzberg MCP server (extractfile/extractbytes), returning a short preliminarynote used to enrich screening/ranking/opportunity-extraction. Falls back grac...
Summary TestHealthConfigValidation in pkg/cli/healthcommandtest.go previously fell through to a real GitHub API call (healthListWorkflowRuns) for its "valid days" test cases, which could take tens of seconds or hang in sandboxed/offline CI environments and caused pkg/cli test timeouts. The fix stubs this call with a mock that returns an empty result, since the test only validates the days parameter and run-listing behavior is covered elsewhere. Change Classification Type: test Scope: pkg/cli Breaking: No Key Changes | File | Change | Impact | |------|--------|--------| | pkg/cli/healthcommandtest.go | Stubs healthListWorkflowRuns with a mock returning (nil, 0, nil), restored via t.Cleanup...
Q Workflow Optimization Report <details> <summary><b>Issues Found</b></summary> Contribution Check (.github/agents/contribution-checker.agent.md) Reported in PR #55049 (comment https://github.com/github/gh-aw/pull/55049#issuecomment-5386168024): the Contribution Check bot posted a "thanks for this tidy PR" style comment on a PR authored by github-actions[bot] (the Instructions Janitor automation), which is internal repo maintenance, not an external contribution. Treating a bot-authored maintenance PR the same as an external contributor's PR is confusing and unnecessary noise. </details> <details> <summary><b>Changes Made</b></summary> Contribution Checker subagent (.github/agents/contribu...
Q Workflow Optimization Report <details> <summary><b>Request</b></summary> Comment on issue #55389 (PR Triage Agent report) requested: "use mai flash model to reduce cost". </details> <details> <summary><b>Changes Made</b></summary> pr-triage-agent (.github/workflows/pr-triage-agent.md) Added model: copilot/mai-code-1-flash-picker to the engine: block. Previously the workflow used the Copilot engine's default model with no explicit model override, which resolves to a more expensive default. copilot/mai-code-1-flash-picker is already used across 30 other lower-cost/high-frequency workflows in this repo (e.g. artifacts-summary.md, issue-arborist.md, metrics-collector.md) as the standard cos...
Workflow Failure Workflow: Detection Stats Daily Branch: main Run: https://github.com/github/gh-aw-threat-detection/actions/runs/32617093541 [!WARNING] AI Credits Rate Limit The Copilot API returned a rate limit response (HTTP 429), but the workflow did not report the explicit AI credits budget-exceeded guardrail signal. Used 35.6 of 1K max. <details> <summary>Tips for reducing rate limit issues</summary> Review the cost optimization guidance. Reduce unnecessary model or tool calls in the prompt. Trim large inputs or excess context that does not change the outcome. Split large tasks across smaller runs when possible. </details> Action Required Assign this issue to an agent to debug and fi...
Workflow Failure Workflow: Smoke AllowOnly Branch: main Run: https://github.com/github/gh-aw-mcpg/actions/runs/32616906798 [!WARNING] AI Credits Rate Limit The Copilot API returned a rate limit response (HTTP 429), but the workflow did not report the explicit AI credits budget-exceeded guardrail signal. Used 27.1 of 1K max. <details> <summary>Tips for reducing rate limit issues</summary> Review the cost optimization guidance. Reduce unnecessary model or tool calls in the prompt. Trim large inputs or excess context that does not change the outcome. Split large tasks across smaller runs when possible. </details> Action Required Assign this issue to an agent to debug and fix the issue. <deta...
Workflow Failure Workflow: Read-Only Stress: docker-sbx runtime Branch: test-improver/mount-policy-coverage-a5c61e81bbc250d5 Run: https://github.com/github/gh-aw-mcpg/actions/runs/32587838195 Pull Request: #11651 [!WARNING] AI Credits Rate Limit The Copilot API returned a rate limit response (HTTP 429), but the workflow did not report the explicit AI credits budget-exceeded guardrail signal. Used 32.5 of 1K max. <details> <summary>Tips for reducing rate limit issues</summary> Review the cost optimization guidance. Reduce unnecessary model or tool calls in the prompt. Trim large inputs or excess context that does not change the outcome. Split large tasks across smaller runs when possible. ...
Workflow Failure Workflow: Read-Only Stress: default runtime Branch: test-improver/mount-policy-coverage-a5c61e81bbc250d5 Run: https://github.com/github/gh-aw-mcpg/actions/runs/32587838150 Pull Request: #11651 [!WARNING] AI Credits Rate Limit The Copilot API returned a rate limit response (HTTP 429), but the workflow did not report the explicit AI credits budget-exceeded guardrail signal. Used 31.1 of 1K max. <details> <summary>Tips for reducing rate limit issues</summary> Review the cost optimization guidance. Reduce unnecessary model or tool calls in the prompt. Trim large inputs or excess context that does not change the outcome. Split large tasks across smaller runs when possible. </d...
Workflow Failure Workflow: Read-Only Stress: gVisor runtime Branch: add-tests-guard-ConfigureGlobalCompilationCache-01548e8eef6a327a Run: https://github.com/github/gh-aw-mcpg/actions/runs/32587684415 Pull Request: #11684 [!WARNING] AI Credits Rate Limit The Copilot API returned a rate limit response (HTTP 429), but the workflow did not report the explicit AI credits budget-exceeded guardrail signal. Used 104.5 of 1K max. <details> <summary>Tips for reducing rate limit issues</summary> Review the cost optimization guidance. Reduce unnecessary model or tool calls in the prompt. Trim large inputs or excess context that does not change the outcome. Split large tasks across smaller runs when p...
Workflow Failure Workflow: Smoke AllowOnly Branch: main Run: https://github.com/github/gh-aw-mcpg/actions/runs/32550553226 [!WARNING] AI Credits Rate Limit The Copilot API returned a rate limit response (HTTP 429), but the workflow did not report the explicit AI credits budget-exceeded guardrail signal. Used 27 of 1K max. <details> <summary>Tips for reducing rate limit issues</summary> Review the cost optimization guidance. Reduce unnecessary model or tool calls in the prompt. Trim large inputs or excess context that does not change the outcome. Split large tasks across smaller runs when possible. </details> Action Required Assign this issue to an agent to debug and fix the issue. <detail...
Agent job 32587838178 failed. [!WARNING] AI Credits Rate Limit The Copilot API returned a rate limit response (HTTP 429), but the workflow did not report the explicit AI credits budget-exceeded guardrail signal. Used 31.9 of 1K max. <details> <summary>Tips for reducing rate limit issues</summary> Review the cost optimization guidance. Reduce unnecessary model or tool calls in the prompt. Trim large inputs or excess context that does not change the outcome. Split large tasks across smaller runs when possible. </details>
QA findings for README Tested at: 2026-08-21T07:32:00Z Viewports: small (360×800), medium (768×1024), desktop (1280×800), wide (1440×900), full-hd (1920×1080) | # | Viewport | Category | Description | |---|----------|----------|-------------| | 1 | small | readability | Horizontal overflow at 360px viewport. A long navigation link (Fix Codespaces actions:write Errors When Running gh aw run) containing an inline <code> element extends to 361px, causing a horizontal scrollbar (scrollWidth 361 > clientWidth 345). The overflowing element right edge reaches 360.8px — outside the 345px client area. | Screenshots !README page at 360px viewport showing horizontal overflow caused by long link with...
Summary The workshop order graph was reviewed across 92 pages (92 nodes, 326 edges). The graph is largely healthy. Three reported cycles and three "weird cycles" are all false positives produced by the heuristic treating bidirectional prerequisite declarations (A lists B as its next step; B lists A as a before-requirement) as loops — these are correct, intentional navigational patterns (error-recovery round-trips, alternative setup paths). One confirmed ordering problem was found. Mermaid graph Ordering constraints (sample — 326 total) 00-welcome.md → 01-prerequisites.md (next link) 01-prerequisites.md → 02a-setup-codespace.md (next link) 02a-setup-codespace.md → side-quest-01-02-environm...
File reviewed workshop/side-quest-13-04-token-optimization.md Violations No standalone "See..." footers — links belong inline Offending text (line 35): Offending text (line 116): Suggested fix: Remove these standalone blockquote lines. Embed the links inline in the preceding prose — for example, change the gh aw audit command reference to use an inline link to the audit reference page, and fold the Cost Management link into the preceding bullet list sentence. [!WARNING] <details> <summary>Firewall blocked 1 domain</summary> The following domain was blocked by the firewall during workflow execution: - awmgmcpg To allow these domains, add them to the network.allowed list in your workflow fr...
Workflow Failure Workflow: Daily Copilot Token Optimization Advisor Branch: main Run: https://github.com/github/gh-aw-firewall/actions/runs/32010582855 [!WARNING] Missing Data Reported: The agent reported missing data during execution. Missing Data: Token usage report issue content: All candidate token-usage-report issues \(github/gh-aw-firewall#7435, #7409, #7408, #7385, #7354\) were filtered by integrity policy as having lower integrity than required, so their bodies could not be read to identify the most token-intensive workflow. - Context: Task requires reading the latest issue labeled token-usage-report to extract per-workflow token stats, but the github MCP tool returned all results...
Workflow Failure Workflow: Refactoring Opportunity Scanner Branch: main Run: https://github.com/github/gh-aw-firewall/actions/runs/31564005359 [!WARNING] AI Credits Rate Limit The Copilot API returned a rate limit response (HTTP 429), but the workflow did not report the explicit AI credits budget-exceeded guardrail signal. Used 37.3 of 1K max. <details> <summary>Tips for reducing rate limit issues</summary> Review the cost optimization guidance. Reduce unnecessary model or tool calls in the prompt. Trim large inputs or excess context that does not change the outcome. Split large tasks across smaller runs when possible. </details> [!WARNING] Missing Tools Reported: The agent reported missi...
Workflow Failure Workflow: Daily Copilot Token Optimization Advisor Branch: main Run: https://github.com/github/gh-aw-firewall/actions/runs/31784935064 [!WARNING] Missing Data Reported: The agent reported missing data during execution. Missing Data: Token usage report issue body \(label: token-usage-report\): All 5 issues found with the token-usage-report label \(including the most recent, #7354\) were filtered out by integrity policy as having 'lower integrity than agent requires \(approved\)', preventing any read of workflow statistics needed to identify the targ \[Content truncated due to length\] - Context: Repository github/gh-aw-firewall, searched via github-list\issues with label t...
Workflow Failure Workflow: Daily Copilot Token Optimization Advisor Branch: main Run: https://github.com/github/gh-aw-firewall/actions/runs/31579473337 [!WARNING] Missing Data Reported: The agent reported missing data during execution. Missing Data: Token usage report issue body \(#7275\): The GitHub MCP server marked issue github/gh-aw-firewall#7275 \(latest token-usage-report\) as filtered by secrecy policy, stating the agent is not authorized to access private-scoped data. Without the report body, per-workflow token/cost statistics cannot b \[Content truncated due to length\] - Context: Step 1 of the Daily Copilot Token Optimization Advisor requires listing/reading the latest issue lab...
Workflow Failure Workflow: Daily Copilot Token Optimization Advisor Branch: main Run: https://github.com/github/gh-aw-firewall/actions/runs/31473816190 [!WARNING] Missing Data Reported: The agent reported missing data during execution. Missing Data: Token usage report issue body content: Cannot extract per-workflow token statistics needed to identify the most token-intensive workflow and generate optimization recommendations - Context: Found candidate token-usage-report issues \(e.g. #7236, #3840, #7143, #7113, #7025, #6983\) via GitHub search/list, but every one is blocked by integrity/secrecy policy: 'Resource has lower integrity than agent requires' and 'not authorized to access privat...
Workflow Failure Workflow: Issue Duplication Detector Branch: main Run: https://github.com/github/gh-aw-firewall/actions/runs/31449330030 [!WARNING] AI Credits Rate Limit The Copilot API returned a rate limit response (HTTP 429), but the workflow did not report the explicit AI credits budget-exceeded guardrail signal. Used 21.6 of 1K max. <details> <summary>Tips for reducing rate limit issues</summary> Review the cost optimization guidance. Reduce unnecessary model or tool calls in the prompt. Trim large inputs or excess context that does not change the outcome. Split large tasks across smaller runs when possible. </details> Action Required Assign this issue to an agent to debug and fix t...
Workflow Failure Workflow: Runner Doctor Updater Branch: main Run: https://github.com/github/gh-aw-firewall/actions/runs/31422860980 [!WARNING] AI Credits Rate Limit The Copilot API returned a rate limit response (HTTP 429), but the workflow did not report the explicit AI credits budget-exceeded guardrail signal. Used 43.9 of 1K max. <details> <summary>Tips for reducing rate limit issues</summary> Review the cost optimization guidance. Reduce unnecessary model or tool calls in the prompt. Trim large inputs or excess context that does not change the outcome. Split large tasks across smaller runs when possible. </details> Action Required Assign this issue to an agent to debug and fix the is...
Workflow Failure Workflow: gh-aw Parity Monitor Branch: main Run: https://github.com/github/gh-aw-threat-detection/actions/runs/31147950679 [!WARNING] AI Credits Rate Limit The Copilot API returned a rate limit response (HTTP 429), but the workflow did not report the explicit AI credits budget-exceeded guardrail signal. Used 21.6 of 1K max. <details> <summary>Tips for reducing rate limit issues</summary> Review the cost optimization guidance. Reduce unnecessary model or tool calls in the prompt. Trim large inputs or excess context that does not change the outcome. Split large tasks across smaller runs when possible. </details> Action Required Assign this issue to an agent to debug and fix...
Workflow Failure Workflow: API Surface & Export Audit Branch: main Run: https://github.com/github/gh-aw-firewall/actions/runs/31374834270 [!WARNING] AI Credits Rate Limit The Copilot API returned a rate limit response (HTTP 429), but the workflow did not report the explicit AI credits budget-exceeded guardrail signal. Used 21.1 of 1K max. <details> <summary>Tips for reducing rate limit issues</summary> Review the cost optimization guidance. Reduce unnecessary model or tool calls in the prompt. Trim large inputs or excess context that does not change the outcome. Split large tasks across smaller runs when possible. </details> Action Required Assign this issue to an agent to debug and fix t...
Workflow Failure Workflow: Daily Copilot Token Optimization Advisor Branch: main Run: https://github.com/github/gh-aw-firewall/actions/runs/31303512943 [!WARNING] Missing Data Reported: The agent reported missing data during execution. Missing Data: Token usage report issue body \(#7143, label token-usage-report\): Need per-workflow token/cost statistics from the latest token-usage-report issue to identify the most token-intensive workflow and build optimization recommendations. - Context: Issue github/gh-aw-firewall#7143 was found via search but the MCP server refused to read it: 'has secrecy requirements that agent doesn't meet / lower integrity than agent requires'. Local pre-agent fil...
Workflow Failure Workflow: Daily Copilot Token Optimization Advisor Branch: main Run: https://github.com/github/gh-aw-firewall/actions/runs/31371559397 [!WARNING] Missing Data Reported: The agent reported missing data during execution. Missing Data: Token usage report issue body: Issue #7143 \(labeled token-usage-report\) was found via search/list, but its content was filtered out by an integrity policy \("Resource has lower integrity than agent requires. The agent cannot read data with integrity below 'approved'"\). Without the report \[Content truncated due to length\] - Context: Repository: github/gh-aw-firewall. Attempted github search\issues and list\issues with label token-usage-rep...
Workflow Failure Workflow: Issue Duplication Detector Branch: main Run: https://github.com/github/gh-aw-firewall/actions/runs/31341481435 [!WARNING] AI Credits Rate Limit The Copilot API returned a rate limit response (HTTP 429), but the workflow did not report the explicit AI credits budget-exceeded guardrail signal. Used 18.4 of 1K max. <details> <summary>Tips for reducing rate limit issues</summary> Review the cost optimization guidance. Reduce unnecessary model or tool calls in the prompt. Trim large inputs or excess context that does not change the outcome. Split large tasks across smaller runs when possible. </details> Action Required Assign this issue to an agent to debug and fix t...
Workflow Failure Workflow: Runner Doctor Updater Branch: main Run: https://github.com/github/gh-aw-firewall/actions/runs/31330734776 [!WARNING] AI Credits Rate Limit The Copilot API returned a rate limit response (HTTP 429), but the workflow did not report the explicit AI credits budget-exceeded guardrail signal. Used 45.8 of 1K max. <details> <summary>Tips for reducing rate limit issues</summary> Review the cost optimization guidance. Reduce unnecessary model or tool calls in the prompt. Trim large inputs or excess context that does not change the outcome. Split large tasks across smaller runs when possible. </details> Action Required Assign this issue to an agent to debug and fix the is...
Workflow Failure Workflow: Daily Copilot Token Optimization Advisor Branch: main Run: https://github.com/github/gh-aw-firewall/actions/runs/31248375829 [!WARNING] Missing Data Reported: The agent reported missing data during execution. Missing Data: Token usage report issue body: The two candidate token-usage-report issues \(#7113 and #3840\) found via search/list are filtered by secrecy/integrity policy and their bodies cannot be read, so per-workflow cost/token stats required by Step 1-2 cannot be extracted. - Context: Repo github/gh-aw-firewall; search\issues and list\issues both returned these issues but issue\read get was blocked: 'secrecy requirements that agent doesn't meet, not au...
Workflow Failure Workflow: Daily Copilot Token Optimization Advisor Branch: main Run: https://github.com/github/gh-aw-firewall/actions/runs/31162404243 [!WARNING] Missing Data Reported: The agent reported missing data during execution. Missing Data: Token usage report issue #7025 body content: Task requires reading the latest token usage report issue \(repo:github/gh-aw-firewall#7025\) to extract per-workflow statistics, but the GitHub MCP tool blocked access with error: 'issue exists but is not accessible — filtered by secrecy policy \(Resource ha \[Content truncated due to length\] - Context: Attempted search\issues \(worked, found #7025\) then issue\read get \(blocked by secrecy policy...
Workflow Failure Workflow: Runner Doctor Updater Branch: main Run: https://github.com/github/gh-aw-firewall/actions/runs/31127146576 [!WARNING] AI Credits Rate Limit The Copilot API returned a rate limit response (HTTP 429), but the workflow did not report the explicit AI credits budget-exceeded guardrail signal. Used 43.5 of 1K max. <details> <summary>Tips for reducing rate limit issues</summary> Review the cost optimization guidance. Reduce unnecessary model or tool calls in the prompt. Trim large inputs or excess context that does not change the outcome. Split large tasks across smaller runs when possible. </details> Action Required Assign this issue to an agent to debug and fix the is...
Workflow Failure Workflow: Daily Copilot Token Optimization Advisor Branch: main Run: https://github.com/github/gh-aw-firewall/actions/runs/31088013925 [!WARNING] Missing Data Reported: The agent reported missing data during execution. Missing Data: Token usage report issue body \(#6983, label token-usage-report\): The workflow requires reading the latest token-usage-report issue body to identify the most token-intensive workflow and its metrics, but issue #6983 was filtered out by the integrity policy \('lower integrity than agent requires. The agent cannot read data \[Content truncated due to length\] - Context: repo github/gh-aw-firewall, issue #6983, label token-usage-report, queried ...
Workflow Failure Workflow: Daily Copilot Token Optimization Advisor Branch: main Run: https://github.com/github/gh-aw-firewall/actions/runs/30895322150 [!WARNING] Missing Data Reported: The agent reported missing data during execution. Missing Data: Token usage report issue content \(#6880\): Task requires reading the latest token-usage-report issue body to identify the most token-intensive workflow, but the GitHub MCP tool blocked access due to secrecy/integrity policy filtering. - Context: github-search\issues and github-issue\read both returned issue #6880 \(label token-usage-report\) but access was denied: 'Resource has secrecy requirements that agent doesn't meet' / 'lower integrity ...
Workflow Failure Workflow: gh-aw Parity Monitor Branch: main Run: https://github.com/github/gh-aw-threat-detection/actions/runs/30879082683 [!WARNING] AI Credits Rate Limit The Copilot API returned a rate limit response (HTTP 429), but the workflow did not report the explicit AI credits budget-exceeded guardrail signal. Used 23.6 of 1K max. <details> <summary>Tips for reducing rate limit issues</summary> Review the cost optimization guidance. Reduce unnecessary model or tool calls in the prompt. Trim large inputs or excess context that does not change the outcome. Split large tasks across smaller runs when possible. </details> Action Required Assign this issue to an agent to debug and fix...
Workflow Failure Workflow: Runner Doctor Updater Branch: main Run: https://github.com/github/gh-aw-firewall/actions/runs/30845118957 [!WARNING] AI Credits Rate Limit The Copilot API returned a rate limit response (HTTP 429), but the workflow did not report the explicit AI credits budget-exceeded guardrail signal. Used 49.2 of 1K max. <details> <summary>Tips for reducing rate limit issues</summary> Review the cost optimization guidance. Reduce unnecessary model or tool calls in the prompt. Trim large inputs or excess context that does not change the outcome. Split large tasks across smaller runs when possible. </details> Action Required Assign this issue to an agent to debug and fix the is...
Workflow Failure Workflow: Daily Copilot Token Optimization Advisor Branch: main Run: https://github.com/github/gh-aw-firewall/actions/runs/30801277360 [!WARNING] Missing Data Reported: The agent reported missing data during execution. Missing Data: Copilot token usage report issue content: Task requires reading the latest token-usage-report issue body to identify the most token-intensive workflow and its metrics, but all matching issues \(#6880, #6818, #6796\) are filtered by secrecy policy as private-scoped data inaccessible to this agent. gh \[Content truncated due to length\] - Context: Attempted github-search\issues and github-list\issues with label:token-usage-report on github/gh-aw...
Workflow Failure Workflow: Runner Doctor Updater Branch: main Run: https://github.com/github/gh-aw-firewall/actions/runs/30762940204 [!WARNING] AI Credits Rate Limit The Copilot API returned a rate limit response (HTTP 429), but the workflow did not report the explicit AI credits budget-exceeded guardrail signal. Used 37 of 1K max. <details> <summary>Tips for reducing rate limit issues</summary> Review the cost optimization guidance. Reduce unnecessary model or tool calls in the prompt. Trim large inputs or excess context that does not change the outcome. Split large tasks across smaller runs when possible. </details> Action Required Assign this issue to an agent to debug and fix the issu...
Workflow Failure Workflow: Daily Copilot Token Optimization Advisor Branch: main Run: https://github.com/github/gh-aw-firewall/actions/runs/30740881412 [!WARNING] Missing Data Reported: The agent reported missing data during execution. Missing Data: Token usage report issue content: The most recent token-usage-report issue \(#6818\) exists but is inaccessible due to secrecy policy filtering, preventing extraction of per-workflow statistics needed to identify the most token-intensive workflow. - Context: Issue github/gh-aw-firewall#6818 found via search but issue\read returned a secrecy-policy filter error: agent not authorized to access private-scoped data. - Alternatives: Re-run with an ...
Workflow Failure Workflow: Daily Copilot Token Optimization Advisor Branch: main Run: https://github.com/github/gh-aw-firewall/actions/runs/30692933631 [!WARNING] Missing Data Reported: The agent reported missing data during execution. Missing Data: Token usage report issue content and workflow run logs: Found candidate token-usage-report issues \(#6818, #6796, #6745, #6716, #6680\) but their content was blocked by integrity policy \(below required 'approved' level\), and the pre-downloaded copilot-logs.json contains zero runs for the last 7 days. Without read \[Content truncated due to length\] - Context: Repo github/gh-aw-firewall; /tmp/gh-aw/token-audit/copilot-logs.json has total\runs...
Workflow Failure Workflow: Daily Copilot Token Optimization Advisor Branch: main Run: https://github.com/github/gh-aw-firewall/actions/runs/30620294848 [!WARNING] Missing Data Reported: The agent reported missing data during execution. Missing Data: Token usage report issue body: Cannot read the latest token-usage-report issue content to identify the most token-intensive workflow - Context: All token-usage-report issues \(e.g. #6796, #6745, #6716\) are filtered by secrecy/integrity policy - agent lacks authorization to access private-scoped or below-approved-integrity data - Alternatives: A maintainer with appropriate access could manually review the latest token-usage-report issue and ru...
File reviewed workshop/14b-pr-reviewer-workflow.md Violations No "See Also" sections — documentation links belong inline Offending text: Suggested fix: Remove the ## Go Further: PR Reviewer Patterns section entirely. This is a de-facto "See Also" hub placed after the checkpoint and Next link. Instead, integrate references to individual side quests as natural inline mentions within the body of the step (for example, when describing the labeler capability, add a parenthetical link to the labeler pattern side quest). [!WARNING] <details> <summary>Firewall blocked 1 domain</summary> The following domain was blocked by the firewall during workflow execution: - awmgmcpg To allow these domains, ...
Part Summary | Part | Files | Mean Score | Std Dev | |---|---|---|---| | Part 1 — core path (lessons 00–14) | 52 | 8.77 / 10.0 | ±0.45 | | Part 2 — advanced (lessons 15+) | 30 | 8.82 / 10.0 | ±0.42 | | Overall corpus | 83 | 8.78 / 10.0 | ±0.44 | Note: The corpus data classifies 52 files as Part 1 (lessons 00–14 plus their side quests) and 30 files as Part 2 (lessons 15+). One file (side-quest-enterprise-setup.md) is categorized as "Other" and appears in a separate table below. Part 1 Scores (lessons 00–14) | File | Overall Score | |---|---| | workshop/05-agentic-workflows-intro.md | 7.78 / 10.0 | | workshop/02a-setup-codespace.md | 8.05 / 10.0 | | workshop/side-quest-09-01f-debugging-chec...
Summary Which workshop file changed workshop/side-quest-01-03-permission-errors.md — the optional side quest that helps learners resolve permission denied errors. What changed This PR creates a new explanatory diagram (not a migration). The file had no images before. SVG files added workshop/images/side-quest-01-03-permission-fix-light.svg workshop/images/side-quest-01-03-permission-fix-dark.svg Both are 1200x560 self-contained SVGs following the workshop light/dark palette. A <picture> block using the canonical theme-aware format was inserted near the How to fix it section, with the light variant as the fallback src. Why the diagram helps learners The permission-error side quest is pure ...
Doc links added (round 8) This PR adds inline hyperlinks for key gh-aw concepts at their first bare occurrence in 4 workshop files. workshop/side-quest-13-04-token-optimization.md AIC → Cost Management — AI Credits (AIC) Added reference to gh aw audit near the audit command Added Cost Management link near guardrail options workshop/side-quest-16-01-github-output.md Frontmatter → Frontmatter reference workshop/side-quest-16-02-secrets-and-permissions.md frontmatter → Frontmatter reference workshop/side-quest-16-05-long-lived-credentials.md permissions → GitHub Tools Read Permissions [!WARNING] <details> <summary>Firewall blocked 1 domain</summary> The following domain was blocked by the fi...
QA findings for 07-your-first-workflow Tested at: 2026-07-22T20:13:38Z Viewports: mobile (375×667), tablet (768×1024), desktop (1280×800) | # | Viewport | Category | Description | |---|----------|----------|-------------| | 1 | mobile | readability | div.workshop-navigation-next (width 430px) extends beyond the .markdown-body container (360px) when navigating directly to the page hash at 375px viewport width, causing a horizontal scrollbar. The element is not constrained by max-width or overflow: hidden on mobile. | Screenshots !Finding 1 — horizontal overflow on mobile (workshop-navigation-next) Detected by Workshop Playwright QA — run #29953629384 [!WARNING] <details> <summary>Firewall ...
The q workflow could only emit PR-based outputs, which blocked publishing optimization summaries as issues when no code changes were appropriate. This update adds an issue output path with bounded creation limits and keeps compiled workflow behavior in sync. Safe output surface update (.github/workflows/q.md) - Added create-issue to safe-outputs with the same routing metadata used by PR output ([q] prefix + automation labels). - Capped issue creation to max: 5 and set expires: 1d. Compiled workflow sync (.github/workflows/q.lock.yml) - Regenerated lockfile so runtime config includes createissue constraints and handler config. Fixes #1765